"Incorporating Entry": a New Approach to Cataloging Electronic Journals

Sara Shatford Layne, May 18, 1998


Should we use "successive entry" for electronic journals? Or should we use "single entry," either in the form of "latest entry" or "earliest entry"? Recent discussions and presentations on these two approaches (most recently by Regina Reynolds at the CONSER Operations Committee Meeting at the Library of Congress) have revealed that neither is truly satisfactory and that there are problems associated with either approach. Electronic journals do not seem to fit cleanly into either the successive or the integrating categories of the Modified Model C developed by Jean Hirons and Regina Reynolds.

In the pre-electronic environment the particular title of a journal issue was permanently connected with that issue (except perhaps in those libraries that discarded the covers of journal issues during the binding process). In the electronic environment, the physical bond between particular issues and particular titles no longer exists. The title of an electronic journal may appear on a "home page" for that journal and nowhere else; this means that when the title on that "home page" changes, it is quite possible that there will be no trace of any previous titles that may have appeared on that home page. A particular issue, and the articles that it contains, will no longer be permanently associated with a particular title. While successive entry preserves the identity of each "title" it also presents a misleading picture of sequential and discrete entities, implying that a particular set of issues is permanently associated with each title. Single entry, although it may present a more accurate picture of the what often appears to be a single entity, can make identification of earlier "titles" quite difficult. What is happening in the electronic environment is that although an earlier "title" may cease to exist as a discrete entity (except in an archived form); its contents continue to exist as a part of the current "title."

In our approach to cataloging electronic journals, let us acknowledge that the earlier "titles" of an electronic journal are parts of the later "titles" of that journal, and treat them as such. One can think of each successive "title" of an electronic journal as incorporating the previous "title," like boxes within boxes. In a sense, the earlier "titles" can be regarded as analytical entries to the later "titles," and we can use an approach that is similar to that used for analytical entries. We can call this new approach "Incorporating Entry."

Let us look at an example of how "incorporating entry" could be implemented, using two "titles": BMMR and The Medieval Review. In May of 1997, the electronic journal previously known as BMMR changed its title to The Medieval Review, and all the issues that previously bore the title BMMR were reformatted with the later title.

Current practice: Successive Entry
[Note: these are edited excerpts from actual records; thanks to John Riemer for identifying them]

022    1070-3616
245 00 BMMR $h [computer file]
362 0  93.8.1-97.05.02.
580    The publication changed title to: Medieval review, and
reformatted all old issues to build a complete archive under the
new title.
785 10 $t Medieval review $x 1096-746X

022    1096-746X
245 04 The medieval review $h [computer file]
362 0  97.05.03-
580    All issues previously published under the former title
BMMR [Bryn Mawr medieval review], have been reformatted with the
new title: The medieval review.
780 10 $t BMMR $x 1070-3616

In this example, using successive entry, the records are linked with USMARC 780 and 785 fields and a cataloger has created notes (USMARC 580) to describe the relationship between the two "titles." The statement of extent (USMARC 362) for the new "title" begins with the issue that was the first to have its original publication under the new title.

Proposed Practice: Incorporating Entry

022    1070-3616
245 00 BMMR $h [computer file]
362 0  93.8.1-97.05.02.
580    Incorporated into: Medieval review.
7?? 10 $t Medieval review $x 1096-746X


022    1096-746X
245 04 The medieval review $h [computer file]
362 0  93.8.1-
580    Incorporates: BMMR.
7?? 10 $t BMMR $g Aug. 1993-May 1997 $x 1070-3616
730 02 BMMR.

In this example, the USMARC tags for the links are not specifically designated. It might be desirable to define new fields (e.g., 788, 789) that are specific to the "Incorporated into/ Incorporates" relationship, as well as a new value for the publication status of an "incorporated" journal. However, the "incorporating entry" approach could be implemented without such changes to the USMARC format. For instance, we might use fields 773 (Host item entry) and 774 (Component item entry) along with the newly defined $I (e.g., 773 $i Incorporates $t BMMR).

Advantages of the Incorporating Entry Approach


          022    1096-746X
          245 04 The medieval review $h [computer file]
          362 0  93.8.1-
          580    Incorporates: BMMR.
          730 02 BMMR.

Questions To Be Resolved:

1) How many previous "titles" are linked to a later "title"? For example, suppose that The Medieval Review changes title once again, this time to The Michigan Medieval Review. Does this new record have links to both The Medieval Review and BMMR?
     022    xxxx-xxxx
     245 04 The Michigan Medieval Review $h computer file
     362 1  93.8.1-
     580    Incorporates: Medieval review; BMMR.
     7?? 10 $t BMMR $g Aug. 1993-May 1997 $x 1070-3616
     7?? 10 $t Medieval review $g {dates} $x 1096-746X

Or does it have just one link, to The Medieval Review, which in its turn is linked to BMMR? It seems that the single link approach would require less maintenance, but it would mean that a cataloging agency that chooses to use just the most current record might lose the BMMR relationship. This will be a matter for CONSER and AACR2 to decide since the national record will require some linkage.

A library could choose to use a current-record only approach and include all the earlier titles using the 730 fields, as below:
     022    xxxx-xxxx
     245 04 The Michigan Medieval Review $h computer file
     362 1  93.8.1-
     580    Incorporates: Medieval review; BMMR.
     730 02 Medieval review.
     730 02 BMMR. 

The amount of editing of the national/CONSER record is a question for institutions to decide.

2) How much information about earlier "titles" should be included in the records for "later" titles? As suggested above, it seems reasonable to suggest that only that information present in the "title" at the time it is cataloged should be included in the record. Information about earlier "titles" can be included in the records for those titles. This is comparable to the current practice for analytical entries: one does not include all the access points for the analytical entry in the record for the entity of which it is a part. If one chooses not to make a separate analytical entry, but rather to make just an analytical added entry on the record for the larger entity, one thereby decides that providing access by "main entry" alone to the smaller entity is sufficient for one's purposes.

3) What happens to holdings? It seems reasonable to suggest that the approach (or approaches) currently taken to recording holdings for existing analytical entries would be appropriate for Incorporating Entries.

4) To what categories of bibliographic entities shown in the Modified Model C should this Incorporating Entry approach apply? It has been developed for application to electronic journals, but might it have a wider application? Could it be applied more broadly to electronic serials, including for example electronic newsletters and annual reports? Could it applied to Integrating Entities such as electronic databases and even paper loose-leaf publications? Applying it more broadly seems to offer many advantages, as these other entities share with electronic journals the characteristic that they incorporate what has gone before, although, in the case of Integrating Entities, the earlier content is fully integrated and does not remain in discrete packages.

5) How does the Incorporating Entry approach affect links between versions of an electronic journal or between that journal and related journals? For example, how does one link the paper version of a journal, cataloged using the successive entry approach, with the electronic journal, cataloged using the Incorporating Entry approach? What happens to the single record approach to cataloging paper and electronic versions of a journal? These questions can be answered, but they will require some thought.

Summary

Incorporating Entry is an intriguing possibility for electronic journals and other types of publications as well. It resolves many of the problems that would be caused by using either successive or latest entry conventions as we now know them. Furthermore, it is in keeping with the spirit of the Hirons/Graham paper because it presents an approach that recognizes the "form of issuance" of the materials to which it would be applied.

Library of Congress

Library of Congress Help Desk (05/28/98)